Photophysics of aqueous tryptophan: pH and temperature effects

Sep 1, 1980 - Peggy A. Thompson, Victor Muñoz, Gouri S. Jas, Eric R. Henry, William A. Eaton, and James Hofrichter. The Journal of Physical Chemistry...
0 downloads 0 Views 1MB Size
J. Am. Chem. SOC.1980, 102, 6271-6279 constant for internal return in 70% dioxane is estimated to be 40 times larger than k4. The fact that k4 is an apparent dissociation rate constant for an oriented complex held together by secondary valence forces makes it impossible to accurately compare k4 to the limiting values for internal return (k-,). The observation of a large isotope effect of 15% does not require that k 4 be much less than k-, since KirbyZohas observed similar effects of 5-16% (20) Craze, G.-A,; Kirby, A. J.; Osborne, R. J . Chem. Soc., Perkin Trans. 2 1978, 357.

627 1

for SN2displacement reactions involving formaldehyde methyl phenyl acetals. The large negative p of -4 that is observed in 70% dioxane suggests that if the reaction is SN2-like, the incoming and leaving groups are widely separated in space. This large amount of effective charge on the central carbon would be expected to have a significant effect on the zero-point energies for the two states under consideration, giving rise to the large isotope effect. If displacement reactions involving acetals commonly exhibit late, loose transition states such as this, large isotope effects such as those that are observed with glycosides would be anticipated.

Photophysics of Aqueous Tryptophan: pH and Temperature Effects R. J. Robbins,? G . R. Fleming,*t G . S. Beddard,' G. W. Robinson,$ P. J. Thistlethwaite,$ and G . J. Woolfe' Contribution from The Davy Faraday Laboratory of the Royal Institution, London W1X 4BS, United Kingdom, and Department of Physical Chemistry, University of Melbourne, Parkville, Victoria 3052, Australia. Received July 26, 1979

Abstract: The fluorescence decay kinetics of aqueous tryptophan and 3-methylindole have been determined as a function of pH and temperature by using a picosecond dye laser-single photon counting system with a time resolution of 50 ps. At pH 11, tryptophan exhibits a single exponential decay, with a lifetime of 9.1 ns at 18 O C . However, at pH 7 the decay is faster and definitely nonexponential; the values obtained from a biexponential fit to the data at pH 7 are T ] = 0.43 ns, 7 = 3.32 ns, a n d f = 0.19 at 18 O C . The behavior of a 3-methylindole closely resembles that of tryptophan at pH 11. A model for the photophysics of aqueous tryptophan is presented in which the excited-state decay constant at pH 11 (where the amino acid side chain is not protonated) is given by the superposition of three independent processes: fluorescence,intersystem crossing, and photoionization; of these processes only photoionization is temperature sensitive (E' = 51 kJ mol-'). In the region pH 4-8, where tryptophan exists in the zwitteridn form, a new nonradiative process is introduced, which involves intramolecular proton transfer from the -NH3+ group to the excited indole ring. The apparent activation energy for intramolecular quenching (E' = 16 kJ mol-') suggests that it is a predominantly diffusion-controlled process. It is proposed that the nonexponential decay observed for aqueous tryptophan at pH 7 arises from transient terms in the rate constant for intramolecular quenching. Quantum yields calculated from this model compare well with experimental values.

Introduction The fluorescence of proteins is usually dominated by that of the tryptophan residues.' Both the fluorescence lifetime and quantum yield of a tryptophan residue are strongly influenced by the nature of its local environment, and this sensitivity is widely exploited through the use of tryptophan as an intrinsic fluorescence probe for the structure and conformation of proteins and polypeptides in solution.2 Interpretation of the results of such experiments requires an understanding of the excited-state decay processes in tryptophan, and their response to environmental perturbations. However, despite extensive investigations3 two fundamental questions remain unanswered. (1) Can an isolated tryptophan residue be characterized by an exponential decay law? (2) What are the principal excited-state decay routes for tryptophan in aqueous solution? Excitation of proteins that contain a few tryptophan residues per molecule yields nonexponential fluorescence decay kinetic^;^,^

the decays can be described as a superposition of a number of exponential components, each of which is assumed to correspond to a tryptophan residue (or group of residues) in a different environment. However, nonexponential decays have also been observed for a number of proteins and peptides containing a single tryptophan r e ~ i d u e ;here, ~ . ~ the nonexponentiality is attributed to the presence of multiple conformations of the molecule. Each of these explanations for nonexponentiality is based on the implicit assumption that the fluorescence decay of an isolated tryptophan residue is strictly exponential. Yet this assumption has not been verified. Recently, Rayner and Szabo8 investigated the fluorescence decay of aqueous tryptophan at pH 7 using the conventional single photon counting technique (excitation by a weak spark lamp of 2-3 ns duration). The fluorescence emission was shown to follow a biexponential decay law with components haviqg lifetimes of 3.14 and 0.51 ns. In a previous paper9 we also reported the observation of a nonexponential fluorescence decay for aqueous

*Address correspondence to this author at the Department of Chemistry and James Franck Institute, The University of Chicago, Chicago, Illinois 60637. Davy Faraday Laboratory. 'University of Melbourne. (1) Teale, F. W. J. Biochem. J . 1960, 76, 381-388. (2) Galley, W. C.; Milton, J. G. Photochem. Photobiol. 1979, 29, 179-184. (3) Weinryb, I.; Steiner, R. F. "Excited States of Proteins and Nucleic Acids", Steiner, R. F., Weinryb, I., Eds.; Plenum Press: New York, 1971; p 271.

(4) Yashinsky, G. Y. Fed. Eur. Biochem. SOC.Lett. 1972, 26, 123-126. (5) Formoso, C.; Forster, L. S. J . Biol. Chem. 1975, 250, 3738-3745. (6) Grinvald, A.; Steinberg, I. Z . Biochim. Biophys. Acta 1976, 427, 663-678. (7) Conti, C.; Forster, L. S. Biochem. Biophys. Res. Commun. 1975,65, 1257-1 263. (8) Rayner, D.M.; Szabo, A. G. Can. J . Chem. 1978, 56, 743-745. (9) Fleming, G. R.; Morris, J. M.; Robbins, R. J.; Woolfe, G. J.; Thistlethwaite, P. J.; Robinson, G. W. Proc. Narl. Acad. Sci. U.S.A. 1978, 75, 4642-4656.

0002-7863/80/1502-6271$01 .OO/O 0 1980 American Chemical Society

6272 J. Am. Chem. Soc., Vol. 102, No. 20, 1980

Robbins et al.

* IOOMHZ

h PD

--

RF SOURCE

Figure 1. Block diagram of the synchronously pumped dye laser and photon counting instrument: ML, mode locker; BF, birefringent tuning element; CD, count-down logic; P, polarizers; PC, Pockels cell; BS, beam splitter; X/2, half-wave rotator; ADA, frequency doubling crystal; F, filters; PD, photodiode; 100 MHz, leading-edge discriminator; CFD, constant fraction discriminator.

S, state. In a previous paper’ we presented a model for the photophysics of aqueous tryptophan in which the excited-state decay was governed by the rates of four independent, competing processes: fluorescence, intersystem crossing, photoionization, and an intramolecular quenching process involving the amino acid side chain. However, because of experimental problems (see below), a quantitative analysis was not possible. In view of these problems and inconsistencies, we have reexamined the fluorescence decay of aqueous tryptophan and the related compound 3-methylindole using a subnanosecond single photon counting apparatus with tunable dye-laser excitation. The results of a systematic study of the fluorescence decay of tryptophan as a function of both pH and temperature complement similar nanosecond laser flash photolysis studies by Bent and Hayon.14 Together these results provide an insight into the origins of nonexponential decays, enable a positive identification of the intramolecular quenching mechanism, and allow a more consistent, quantitative analysis of our previously reported model for the photophysics of aqueous tryptophan.

Experimental Section

A. Chemicals. Tryptophan and 3-methylindole were obtained from Sigma and used without further purification. Samples were stored at 0 tryptophan at pH 7. Using picosecond pulses at 264 nm for OC in a dessicator in the dark. Solutions were prepared in Analar water immediately before use, and the appropriate p H was obtained using HCI excitation and an ultrafast streak camera for detection, we obtained (pH 1,3), BDH buffer tablets (pH 7, 9.2), or N a O H (pH 11, 13). The a biexponential decay with components having lifetimes of 2.1 purity of the samples was checked by a comparison of their absorption, and 5.4 ns. These results are clearly inconsistent with those emission, and excitation spectra. obtained by Rayner and Szabo.8 B. Fluorescence Lifetime Measurements. Fluorescence lifetimes were The use of tryptophan as an intrinsic fluorescence probe requires measured with the use of the single photon counting apparatus shown in an understanding of the possible excited-state decay routes and Figure I . The laser excitation source consisted of a C W rhodamine 6G their dependence on solvent, pH, temperature, and the pressure dye laser (Coherent Radiation CR590) synchronously pumped by an of potential quenchers. Both photophysical and photochemical actively mode-locked argon ion laser (Coherent Radiation CR12). The processes must be considered. They include argon ion laser produced a highly stable train of 90-ps pulses a t a repetition frequency of 75.525 MHz with an average power of 800 m W while intersystem c r o s ~ i n g , photoi~nization,’~-*~ ~~-~~ photodissociation (indole N-H bond and intramolecular q~enching,~’-~* the dye laser, which had a three-plate birefringent tuning element and and, in proteins and peptides, singlet-singlet energy t r a n ~ f e r ~ ~ . a~ 30% ~ transmitting output mirror, produced pulses of